In the spirit of adventure, or my sapless version of it, I asked ChatGPT 3.5 to propose some puns to deploy in the title of this post. The results ranged from the dull to the surreal. (And some results were both dull and surreal… “Between a Rock and a Tax Place” …?!)

Hence my resorting to a hand-crafted title. “Firm Foundations”: it conveys the sense of relief and solidity I felt when reading Lady Justice Falk’s leading judgment in the BlackRock case. 

Whilst the outcome may be easier to see in hindsight than it was in prospect (more on that at the end…), the Court of Appeal’s judgment in BlackRock reflects a depth of understanding that has brought much-needed clarity to the two issues raised in the case.

The Past

First, a quick recap. BlackRock acquired Barclays’ U.S. investment management business in 2009, using an acquisition structure that entailed a Delaware-incorporated but UK-resident company (LLC5) borrowing $4 billion from its US-resident parent (LLC4) and contributing the funds to another US-resident (LLC6), which acquired the target business.

Why was LLC5 resident in the UK — indeed, why did it exist at all? The basis for HMRC’s objection to the structure, in essence, was that LLC5 existed only to harvest corporation tax deductions in respect of interest on the acquisition financing, thus generating tax losses that could be surrendered to profitable UK companies elsewhere in the group.

HMRC attacked the structure on two bases: (i) transfer pricing, and (ii) the unallowable purpose rule which is now to be found in section 441 of the Corporation Tax Act 2009.

The FTT held in favour of the taxpayer on both grounds. The Upper Tribunal (UT) then overturned that decision, holding in favour of HMRC on both issues. Now, the Court of Appeal (CA) has sided with the taxpayer on transfer pricing, but with HMRC on the unallowable purpose rule — meaning that the taxpayer lost the case.

Covenants and Comparability

For regulatory and compliance reasons, the voting rights had been structured such that LLC4 controlled LLC6, but LLC5 did not. In other words, the lender had the ability to control the flow of dividends to the borrower (on which the borrower relied to service the debt), but the borrower did not itself control that dividend flow. Therefore, HMRC argued, an independent lender would not have entered into the loans at all.

In the FTT and UT, this issue had been couched in terms of whether additional covenants to protect the dividend flow could induce a third party to lend. The FTT thought so, and found that, in a third-party context, LLC6 and the target group would have given the necessary covenants (hence, why the taxpayer won on transfer pricing). But the UT considered that, as a matter of transfer pricing law, such covenants could not be hypothesised unless they were included in the intra-group transaction (which they hadn’t been because including such covenants would have been unnecessary and artificial given that LLC4 in fact already controlled the dividend flow).

In my blog post about the UT’s decision, I remarked that that seemed a strange outcome, given that the function of the covenants in the hypothesis was simply to make the risks, assets and functions comparable as between the hypothetical arm’s length transaction and the real one. The Court of Appeal has made exactly that point, holding that nothing in the UK legislation or OECD guidelines requires the position of third parties to be ignored; assuming the existence of the covenants in the hypothetical transaction is part and parcel of ensuring a fair comparison.

That result should offer considerable comfort to taxpayers and advisers. 

It is interesting to note, though, that Lady Justice Falk also hinted (at paragraph 63 of her judgment) that it may have been more pertinent to focus on the borrower’s position than the lender’s — as one can equally ask whether an independent borrower would be prepared to enter into the loans without assurances that its rights in respect of LLC6 would not be frustrated.

Consequences versus Purposes

Having sided with the taxpayer on transfer pricing, the Court of Appeal then turned to the “unallowable purpose” rule, which denies deductions for accounting debits if they are, on a just and reasonable apportionment, attributable to a main purpose of UK tax avoidance.

It was common ground that, following Brebner and TDS, the test of purpose is a matter of subjective intention. This was not a case like Development Securities, in which the board of directors was found effectively to have acted as the puppet of the wider group. Here, HMRC accepted that the board had not been bypassed. The purpose test therefore turned upon the intention of the directors of LLC5 when deciding to enter into the loans: what object or objects were they seeking to achieve?

The Court of Appeal clarified and distilled several principles from the relevant case law:

• Ascertaining the object or purpose of something normally involves an inquiry into the subjective intentions of the relevant actor.
• Object or purpose must be distinguished from effect. Effects or consequences, even if inevitable, are not necessarily the same as objects or purposes.
• Subjective intentions are not limited to conscious motives, and motives are not necessarily the same as objects or purposes.
• “Some” results or consequences are “so inevitably and inextricably involved” in an activity that, unless they are merely incidental, they must be a purpose for it.

Applying those principles, the CA concluded that both the FTT and the UT had erred in law in determining the existence of a main tax purpose by focusing (respectively) on the consequences of borrowing the funds and the reasons for LLC5’s existence, rather than, as correctly framed, the company’s own subjective purposes for being a party to the loans.

But the outcome for the taxpayer remained the same. Lady Justice Falk considered it a useful starting point to ask: why did LLC5 enter into the loans? And there was an obvious answer: to obtain a tax advantage. The decision of LLC5’s board did not take place in a vacuum, and the wider context could not be artificially divorced from it. She did not overlook that LLC5’s board considered the transaction commercially beneficial for LLC5 on a standalone basis (given it made a margin between dividend inflow and interest outflow). 

The CA remade the FTT’s decision, finding as fact that LLC5 had both a commercial purpose and a main purpose of tax avoidance in relation to the loans.

Which Debits Fit which Purpose?

That meant the debits had to be apportioned between the commercial and the unallowable tax avoidance purpose. 

This is an objective exercise. The legislation does not prescribe the precise mechanism, and counsel for the taxpayer offered three different options:

• A causation-based approach using a but-for test (which was intended to support the FTT’s conclusion that 100% of the debits should be apportioned to the allowable commercial purpose). 
• Apportionment based on the “relative anticipated financial significance” of the interest deductions (taking into account expected disallowances under a different set of rules not at issue in the case) and the margin earned by LLC5. 
• Allocating to the allowable commercial purpose the amount of interest deductions that the taxpayer’s group had expected to be disallowed under those other rules.

The CA agreed with the UT that 100% of the debits should be apportioned to the unallowable purpose and disallowed. The transaction had been presented to LLC5’s board as, effectively, a “fait accompli”, and the commercial advantage to LLC5 was “more in the nature of a by-product” when set in the context of the tax structuring. 

The question of how to apportion the debits to the respective purposes could not be answered by asking (as the FTT had) whether the transaction would nonetheless have gone ahead if the tax relief had been withdrawn at the last minute (not least because that would be a purely counterfactual exercise). But, in the appropriate case, there may be reason to deploy one of the other two approaches to apportionment suggested by the taxpayer; Lady Justice Falk could “see that some form of apportionment based on economic advantage could be appropriate in some cases”, but this was not one of those cases. 

The Future

It remains to be seen whether or not the parties will seek permission to appeal. Given the quality of the Court of Appeal’s analysis, though, it seems unlikely that the Supreme Court would reach a different conclusion.

It should also be acknowledged, though, that the outcome of the case does not accord with the view that, I suspect, many advisers would have taken if they had been asked to opine on the structure at the time of its inception. In 2009, it would have seemed relatively uncontroversial to contend that the unallowable purpose rule was not intended to go so far as to disallow the deductibility of debt financing for a genuine third-party acquisition, in circumstances where UK tax residence had been chosen for its favourable debt regime.

So, the broader point is that legal norms, judicial attitudes, HMRC’s risk assessments, and the mood music among practitioners, all change with the times. That is one of the most challenging aspects of giving legal advice. We must advise on the law as it stands, but we can also add real value by sensing and communicating the direction of travel.